Central insights gained from the first cycle of AG041851 were that the combination of elevated amyloidosis and neurodegeneration greatly increased risk of clinical progression among both clinical normal and mildly impaired persons; and, a novel finding was the definition of an abnormal biomarker category characterized by positive neurodegeneration/neuronal injury biomarkers in the absence of amyloid. We labeled this category suspected non-Alzheimers pathophysiology (SNAP) on the assumption that it represented common non-AD pathologies - e.g., cerebrovascular disease, Lewy body disease, etc. Two modern diagnostic classification systems exist for Alzheimers disease (AD); the National Institute on Aging-Alzheimers Association (NIA-AA) and the International Working Group (IWG). SNAP is not addressed by either of these criteria, yet roughly of elderly clinically normal (CN) and mild cognitive impairment (MCI) persons fall into this category. In addition, neither the NIA-AA nor the IWG criteria include tau PET for classification. We recently led a large international group of senior investigators in proposing a new, descriptive classification scheme for AD biomarkers (Appendix). The seven most widely regarded AD biomarkers are used to create a 3-class biomarker classification scheme called the ATN system. In the ATN system, amyloid biomarkers are amyloid PET and CSF A?42 (denoted by A); tau biomarkers are tau PET and CSF p tau (denoted by T); and, neurodegeneration/neuronal injury biomarkers are FDG PET, anatomic MRI, and CSF t tau (denoted by N). Individuals are classified as positive or negative in each of the three categories leading to eight possible biomarker states (e.g., A-T-N-, A+T+N-, etc.). Neither the NIA-AA nor the IWG criteria categorize individuals in this way and neither addresses the implications of the eight different ATN biomarker permutations. The aims of this renewal grant will focus on understanding the implications of categorizing individuals into these eight ATN classes. We will use amyloid PET to define A, tau PET to define T, and MRI cortical thickness to define N. Given the current emphasis on individuals who are clinically asymptomatic or have very early signs of cognitive impairment, we will concentrate on individuals who are CN and MCI at baseline. Our aims are: Aim 1: To create fully imaged population-based cohorts of CN and MCI individuals aged 30?90 with baseline amyloid PET, tau PET, and MRI studies who will be followed clinically with visits every 15 months. Aim 2: To determine how clinical and demographic characteristics (e.g., age, sex, APOE, indicators of cerebrovascular disease, and baseline cognitive performance) vary across the eight ATN biomarker states. Aim 3: To estimate the age and sex specific prevalence rates of the eight ATN biomarker states. Aim 4: To determine the associations between the eight ATN biomarker states and cognitive or clinical outcomes and whether covariates (e.g., age, sex, APOE, and cerebrovascular disease) modify rates of cognitive decline.